Scanning Framework Using Neural Radiance Fields for Dynamic Space Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing scanning technologies for generating virtual representations of spaces are static and time-consuming, leading to discrepancies between real and virtual spaces as objects move or change position, and require user intervention for scanning, which is not user-friendly.
Innovation Solution
A scanning framework that continuously maps a real-world space to a virtual representation using a neural radiance field model, passively collecting image data to update the model with semantic difference computation methods, reducing the need for explicit user scanning and minimizing discrepancies through dynamic updates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static scanning is used to generate virtual representations, then the initial model can be created, but discrepancies between real and virtual spaces increase over time as objects move or change position
Solution Approach 1:
The patent implements dynamic scanning where the system continuously or periodically rescans the physical space to detect changes in object positions and updates the virtual representation accordingly. This transforms the static scanning approach into a dynamic one that adapts to real-time changes, resolving the contradiction between maintaining accuracy and avoiding excessive scanning time by only scanning when changes are detected or at optimized intervals
Solution Approach 2:
The system incorporates feedback mechanisms where the virtual representation is continuously compared with new scan data, and discrepancies trigger automated updates. This feedback loop ensures the virtual model remains synchronized with the physical space without requiring constant full rescanning, thus maintaining reliability while minimizing time loss
2Reliability
If continuous scanning is performed to maintain accurate virtual representations, then discrepancies are minimized, but the scanning process becomes time-consuming and not user-friendly
Solution Approach 1:
The system performs self-service by automatically detecting when scanning is needed through change detection algorithms and executing scans only when necessary. This eliminates the need for user intervention to initiate scanning, making the process user-friendly while maintaining accuracy through automated, context-aware scanning decisions
Solution Approach 2:
The scanning frequency and intensity are dynamically adjusted based on detected changes in the physical environment. When objects move or space configuration changes, the system increases scanning activity; when stable, it reduces scanning. This dynamic adaptation maintains reliability while optimizing ease of operation by eliminating unnecessary scanning operations
3Productivity
If manual user intervention is required for scanning, then scanning can be initiated, but the process becomes complex and not user-friendly
Solution Approach 1:
The system is designed to operate autonomously without requiring user initiation or intervention. It automatically monitors the physical space, detects changes, triggers scans when needed, and updates the virtual representation seamlessly in the background, thereby maintaining high productivity while maximizing ease of operation through complete automation
Solution Approach 2:
The patent replaces manual mechanical scanning operations with automated sensor-based detection and computational processing. Imaging devices, sensors, and algorithms substitute for user-driven scanning actions, enabling fast, automated model generation while eliminating the complexity of manual operation
Data Source
AI summary
Disclosed implementations generate a virtual representation of a space based on a model. The model is updated with image data according to a difference metric. The difference metric is determined for a portion of the space based on the image data and a current state of the model. The virtual representation is provided to a user device.